Project summary: Current bioanalysis largely relies on tissue homogenization, separation, followed by various assays. These in vitro approaches tell the presence of molecules and their concentrations. Nevertheless, without spatial and temporal dynamics information, how molecules execute their functions in a living system remains unknown. Moreover, important target molecules with small quantities are often buried in the large background of dominant species in an in vitro assay. Chemical microscopy allows for label-free analysis of biomolecules in their natural environment. Based on molecular fingerprint vibration, recently developed coherent Raman scattering microscopy has enabled a broad spectrum of biomedical applications. Yet, broader use of coherent Raman scattering microscopy for single cell analysis is blocked by the extremely small cross section of Raman scattering (~10-30 cm2. In response to PAR-17-045, we propose to develop a new chemical imaging platform that exceeds the detection sensitivity limit of coherent Raman microscopy. We employ mid-infrared absorption (with a much larger cross-section ~10-22 cm2) as a contrast mechanism for imaging living cells. To bypass the water absorption obstacle, we indirectly detect the thermal effect induced by vibrational absorption. Our scientific premise is that after the mid-infrared photons induce the molecule to vibrate, the subsequent vibrational relaxation into heat causes a local change of the refractive index. Such change creates a phase delay and a thermal lens, both of which can be detected at sub-micron spatial resolution by a visible probe beam. In a proof of concept study in Science Advances (2016, 2: e1600521), we have reached an imaging speed of 8 seconds per frame, a lateral resolution of 0.6 micrometer, and a detection sensitivity level of 10 micro-molar for the endogenous C=O bond. In the proposed study, we will further improve the detection sensitivity by using better laser source and lab-built electronics (Aim 1). Moreover, we will improve the speed to 100 frames per second through wide-field illumination and quantitative phase detection (Aim 2). Finally, we will demonstrate volumetric chemical imaging through a light-sheet scheme (Aim 3). By accomplishing these Aims, we will have produced a highly sensitive, high-speed chemical imaging platform able to map drug distribution in a pharmaceutical formulation, metabolic activities inside a living cell, and membrane potential in a living neuron. These capacities will generate a profound impact on our understanding of life at the molecular level and on biomarker-based precise staging of diseases.